Training in progress, epoch 0

config.json

@@ -1,12 +1,10 @@
 {
-  "_name_or_path": "facebook/timesformer-base-finetuned-k400",
+  "_name_or_path": "google/vivit-b-16x2-kinetics400",
   "architectures": [
-    "TimesformerForVideoClassification"
+    "VivitForVideoClassification"
   ],
   "attention_probs_dropout_prob": 0.0,
-  "attention_type": "divided_space_time",
-  "drop_path_rate": 0,
-  "hidden_act": "gelu",
+  "hidden_act": "gelu_fast",
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
@@ -21,14 +19,22 @@
     "Normal": 1
   },
   "layer_norm_eps": 1e-06,
-  "model_type": "timesformer",
+  "model_type": "vivit",
   "num_attention_heads": 12,
   "num_channels": 3,
-  "num_frames": 8,
+  "num_frames": 32,
   "num_hidden_layers": 12,
-  "patch_size": 16,
-  "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.46.2"
+  "transformers_version": "4.46.2",
+  "tubelet_size": [
+    2,
+    16,
+    16
+  ],
+  "video_size": [
+    32,
+    224,
+    224
+  ]
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:903608ecd4633f039e8b8b6a2d5ea37147691e809cd0e8690c1097787220fcab
-size 485072256
+oid sha256:9e88fb678eba599b8e3b259fe292d101874d8883548e6cb719bd452a2842e93b
+size 354615376

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1805980ec3cb2dd98d906644744c72e815b7396d114e166da66365f259eb46ba
+oid sha256:e0cbba7a357077adc7cee6898192854bb85ec42ee13082c08f384f8aa700a5c9
 size 5304